1. compare
- Posted by Greg Phillips <i.shoot at REDNECKS.COM>
Dec 15, 1998
-
Last edited Dec 16, 1998
Okay, a quick question.... I decided to write a quick program that'd
compare a file, byte by byte, and tell you if they're the same or
different. My program opened two files in read only mode, got the data
from the file, compare()'d it, and returned a "Files are the same" or
"Files are different" string, which worked great, as long as the files
were fairly small....if the files were 10k, I could change one
character, and it would detect the difference, but a 50k file, and I'd
have to change a few characters for it to tell the difference. The
bigger the file, the more char's had to be changed to return a false
value.... hrm... damnit, all I wanted was something like the old dos
compare program....*sigh* oh well, guess I'll have to get those old dos
5 disks out. I really hope they aren't on 5 1/4.
Anyways, this is kinda bugging me, and I'd really like to feel that
familiar pride of writing something myself, and actually using it for
something, too =)
thanks,
Greg
--
Greg Phillips
i.shoot at rednecks.com
http://euphoria.server101.com
--
Useless fact of the day:
Sex burns 360 calories per hour
2. Re: compare
Greg Phillips writes:
> I decided to write a quick program that'd
> compare a file, byte by byte, and tell you if they're the same or
> different. My program opened two files in read only mode, got the data
> from the file, compare()'d it, and returned a "Files are the same" or
> "Files are different" string, which worked great, as long as the files
> were fairly small....if the files were 10k, I could change one
> character, and it would detect the difference, but a 50k file, and I'd
> have to change a few characters for it to tell the difference.
It sounds like your program opens the files with "r" mode (text),
when it should really use "rb" mode (binary). DOS does funny
things when you open a file in text mode and the file has certain
non-text characters in it. For example, when the ASCII code 26
(control-Z) is encountered, DOS will report "end of file", even when
there are lots of characters remaining in the file. Maybe you compared
two files that were the same up to 10K, and then both of them had
26 - they'd appear to be identical, even though they differ
at some point after 10K.
Regards,
Rob Craig
Rapid Deployment Software
http://members.aol.com/FilesEu/
3. Re: compare
Greg Phillips wrote:
>
> Okay, a quick question.... I decided to write a quick program that'd
> compare a file, byte by byte, and tell you if they're the same or
> different. My program opened two files in read only mode, got the data
> from the file, compare()'d it, and returned a "Files are the same" or
> "Files are different" string, which worked great, as long as the files
> were fairly small....if the files were 10k, I could change one
> character, and it would detect the difference, but a 50k file, and I'd
> have to change a few characters for it to tell the difference. The
> bigger the file, the more char's had to be changed to return a false
> value.... hrm... damnit, all I wanted was something like the old dos
> compare program....*sigh* oh well, guess I'll have to get those old dos
> 5 disks out. I really hope they aren't on 5 1/4.
>
> Anyways, this is kinda bugging me, and I'd really like to feel that
> familiar pride of writing something myself, and actually using it for
> something, too =)
>
> thanks,
> Greg
>
> --
> Greg Phillips
> i.shoot at rednecks.com
> http://euphoria.server101.com
> --
>
> Useless fact of the day:
>
> Sex burns 360 calories per hour
Greg, why do you use compare() instead of getc()? getc()
is fast, and you are able to compare files as large as they
may be (not limited by memory). If you want to testify that
files are identical (name and content) look at my DIRCMP.EX
which you may find in the 'user contributions'.
Have a nice day, Rolf
the length of the files
